In recent years, high-definition television (HDTV) broadcasting has been provided, and a display apparatus capable of displaying a video having a resolution of 1920 horizontal pixels×1080 vertical pixels has been commercialized. Hereinafter, the video having a resolution of 1920×1080 is referred to as a Full HD (Full High Definition) video. In addition, as new high-definition video standards, standards of QFHD (Quad Full High Definition) (commonly called 4 k) having four times the number of pixels (twice the number of vertical pixels×twice the number of horizontal pixels) of Full HD and SHV (Super Hi-Vision, or UHDTV (Ultra High Definition Television)) (commonly called 8 k) having sixteen times the number of pixels (four times the number of vertical pixels×four times the number of horizontal pixels) of Full HD have been proposed. For example, when such a high-resolution (high-definition) video is processed, dividing a display screen into a plurality of areas and providing a video processing circuit for each area for parallel processing is effective.
For example, in PTL 1, a configuration is disclosed in which a two-dimensional image of each frame is divided into a plurality of areas and respective signal processing circuits for the plurality of areas divided perform image processing in parallel. Here, in order for each signal processing circuit to independently perform two-dimensional processing, such as spatial filtering, a video signal of a small area (margin) adjacent to the corresponding divided area is input to each signal processing circuit. In addition, each signal processing circuit performs processing such as spatial filtering on the corresponding divided area by using a video signal of the corresponding divided area and the video signal of the margin.
In addition, in PTL 2, a technique of performing frame rate conversion (FRC) processing to generate an interpolated frame image from preceding and succeeding frame images is disclosed. In FRC, a motion vector that represents a motion of an object or a background in a video needs to be obtained.